The .beta.-sheet conformation (also referred to as a .beta.-strand conformation) is a secondary structure present in many polypeptides. The .beta.-sheet conformation is nearly fully extended, with axial distances between adjacent amino acids of approximately 3.5 .ANG.. The .beta.-sheet is stabilized by hydrogen bonds between NH and CO groups in different polypeptides strands. Additionally, the dipoles of the peptide bonds alternate along the strands which imparts intrinsic stability to the .beta.-sheet. The adjacent strands in the .beta.-sheet can run in the same direction (i.e., a parallel .beta.-sheet) or in opposite directions (i.e., an antiparallel .beta.-sheet). Although the two forms differ slightly in dihedral angles, both are sterically favorable. The extended conformation of the .beta.-sheet conformation results in the amino acid side chains protruding on alternating faces of the .beta.-sheet.
The importance of .beta.-sheets in peptides and proteins is well established (e.g., Richardson, Nature 268:495-499, 1977; Halverson et al., J. Am. Chem Soc. 113:6701-6704, 1991; Zhang, J. Biol. Chem. 266:15591-15596, 1991; Madden et al., Nature 353:321-325, 1991). The .beta.-sheet is important in a number of biological protein-protein recognition events, including interactions between proteases and their substrates, protein kinases and their substrates or inhibitors, the binding of SH2 domain containing proteins to their cognate phosphotyrosine containing protein targets, farnesyl transferase to its protein substrates, and MHC I and II and their antigenic peptides, and has been implicated in many disease states.
Inhibitors that mimic the .beta.-sheet structure of biologically active proteins or peptides would have utility in the treatment of a wide variety of conditions. For example, Ras, the protein product of the ras oncogene, is a membrane bound protein involved in signal transduction regulating cell division and growth. Mutations in the ras gene are among the most common genetic abnormalities associated with human cancers (Barbacid, M. "ras genes," 56:779-827, 1987). These mutations result in a growth signal which is always "on," leading to a cancerous cell. In order to localize to the cell membrane, Ras requires prenylation of the cysteine within its C-terminal CaaX sequence by farnesyl transferase (FTase). (In the sequence CaaX "a" is defined as an amino acid with a hydrophobic side chain and "X" is another amino acid.) This post-translational modification is crucial to its activity. Peptidyl inhibitors of FTase with the sequence CaaX have been shown to block or slow the growth of tumors in cell culture and in whole animals (Kohl et al., "Selective inhibition of ras-dependent transformation by a farnesyltransferase inhibitor," Science 260:1934-1937, 1993; Buss, J. E. & Marsters, Jr., J. C. "Farnesyl transferase inhibitors: the successes and surprises of a new class of potential cancer chemotherapeutics," Chemistry and Biology 2:787-791, 1995).
SH2 domains, originally identified in the src subfamily of PTKs, are noncatalytic sequences and consist of about 100 amino acids conserved among a variety of signal transducing proteins (Cohen et al., Cell 80:237-248, 1995). SH2 domains function as phosphotyrosine-binding modules and mediate critical protein-protein associations (Pawson, Nature 573-580, 1995). In particular, the role of SH2 domains has been clearly defined as critical signal transducers for receptor tyrosine kinases (RTKs such as EGF-R, PDGF, insulin receptor, etc.). Phosphotyrosine-containing sites on autophosphorylated RTKs serve as binding sites for SH2-proteins and thereby mediate the activation of biochemical signaling pathways (Carpenter, G., FAESEB J. 6:3283-3289, 1992; Sierke, S. and Koland, J., Biochem. 32:10102-10108, 1993). The SH2 domains are responsible for coupling the activated growth-factor receptors to cellular responses which include alterations in gene expression, cell proliferation, cytoskeletal architecture and metabolism.
At least 20 cytosolic proteins have been identified that contain SH2 domains and function in intracellular signaling. The distribution of SH2 domains is not restricted to a particular protein family, but is found in several classes of proteins, protein kinases, lipid kinases, protein phosphatases, phospholipases, Ras-controlling proteins and some transcription factors. Many of the SH2-containing proteins have known enzymatic activities while others (Grb2 and Crk) function as "linkers" and "adapters" between cell surface receptors and downstream effector molecules (Marengere, L., et al., Nature 369:502-505, 1994). Examples of proteins containing SH2 domains with enzymatic activities that are activated in signal transduction include, but are not limited to, the src subfamily of protein tyrosine kinases (src (pp60.sup.c-src) abl, lck, fyn, fgr and others), phospholipase-C-.gamma. (PLC-.gamma.), phosphatidylinositol 3-kinase (Pl-3-kinase), p21-ras GTPase activating protein (GAP) and SH2 containing protein tyrosine phosphatases (SH-PTPase) (Songyang et al., Cell 72:767-778, 1993). Intracellular tyrosines are phosphorylated when surface receptors are engaged by diverse ligands for growth factor receptors, cytokine receptors, insulin receptor, and antigen-mediated signaling through T- or B-cell receptors. The phosphorylation of proteins at tyrosine residues is critical in the cellular signal transduction, neoplastic transformation and control of the cell cycle. Due to the central role these various SH2-proteins occupy in transmitting signals from activated cell surface receptors into a cascade of additional molecular interactions that ultimately define cellular responses, inhibitors which block specific SH2-protein binding are desirable as agents for a variety of potential therapeutic applications.
Disease areas in which tyrosine phosphorylation and inhibition of SH2 binding represent targets for drug development include the following:
Cancer: SH2 domains which mediate signaling are clearly significant elements in the regulation of oncogene and protooncogene tyrosine kinase activity and cellular proliferation (Carpenter, Fed. Am. Soc. Exp. Biol. J. 6:3283-3289, 1992). The SH2 domains define an important set of substrates through which activated RTKs mediate signaling and through which nonreceptor tyrosine kinases associate with RTKs and are thus targets for anticancer drug development. The ability to block interaction of the RTK with the SH2-containing substrate using a mimetic inhibitor provides a means to abrogate signaling and thereby eliminate oncogenic activity. The biological significance is also illustrated by the v-crk oncogene, a protein composed almost entirely of SH domains, which is able to bring about cellular transformation by interacting with phosphotyrosine containing proteins. As above, the ability of inhibitors to block v-crk binding via its SH2 domain to other proteins would be expected to be effective as an anticancer agent.
Immune Regulation: Regulation of many immune responses is mediated through receptors that transmit signals through tyrosine kinases containing SH2 domains. T-cell activation via the antigen specific T-cell receptor (TCR) initiates a signal transduction cascade leading to lymphokine secretion and cell proliferation. One of the earliest biochemical responses following TCR activation is an increase in tyrosine kinase activity. In particular, T-cell activation and proliferation is controlled through T-cell receptor mediated activation of p56.sup.lck and p59.sup.fyn tyrosine kinases, as well as ZAP-70 and Syk (Weiss and Litman, Cell 76:263-274, 1994) which contain SH2 domains. Additional evidence indicates that several src-family kinases (lck, blk, fyn) participate in signal transduction pathways leading from B-cell antigen receptors and hence may serve to integrate stimuli received from several independent receptor structures. Thus, inhibitors that block interactions of these SH2 domain kinases with their cognate receptors could serve as immunosuppressive agents with utility in autoimmune diseases, transplant rejection or as anti-inflammatory agents as well as anticancer drugs in cases of lymphocytic leukemias.
Additionally, non-transmembrane PTPase containing SH2 domains are known and nomenclature refers to them as SH-PTP1 and SH-PTP2 (Neel, Cell Biology 4:419-432, 1993) SH-PTP1 is identical to PTP1C, HCP or SHP and SH-PTP2 is also known as PTP1D or PTP2C. SH-PTP1 is expressed at high levels in hematopoietic cells of all lineages and all stages of differentiation. Since the SH-PTP1 gene was identified as responsible for the motheaten (me) mouse phenotype, this provides a basis for predicting the effects of inhibitors that would block its interaction with its cellular substates. Thus, inhibition of SH-PTP1 function would be expected to result in impaired T-cell responses to mitogenic stimulation, decreased NK cell function, and depletion of B-cell precursors with potential therapeutic applications as described above.
Diabetes: In Type 2 (non-insulin dependent) diabetes, tyrosine phosphatases (SH-PTP2) counter-balance the effect of activated insulin-receptor kinases and may represent important drug targets. In vitro experiments show that injection of PTPase blocks insulin stimulated-phosphorylation of tyrosyl residues on endogenous proteins. Thus, inhibitors could serve to modulate insulin action in diabetes.
Neural Regeneration: Glial growth factors are ligands that are specific activators of erb-B2 receptor tyrosine kinase (p185.sup.erbB2) to promote tyrosine phosphorylation and mitogenic responses of Schwann cells. Consequently, regulation of tyrosine phosphorylation by altering activity in Schwann cells following nerve injury could be an important therapeutic strategy. Inhibitors of erb-B2 signaling activity could have a significant role in treatment of tumors of glial cell origin.
Another class of .beta.-sheet mimetics are inhibitors of protein kinases, which include the protein tyrosine kinases and serine/threonine kinases.
A wide variety of cellular substrates for polypeptide growth factor receptors that possess intrinsic tyrosine kinase activity have now been characterized. Although there is a tremendous diversity among the numerous members of the receptors tyrosine-kinases (RTK) family, the signaling mechanisms used by these receptors share many common features. Biochemical and molecular genetic studies have shown that binding of the ligand to the extracellular domain of the RTK rapidly activates the intrinsic tyrosine kinase catalytic activity of the intracellular domain. The increased activity results in tyrosine-specific phosphorylation of a number of intracellular substrates which contain a common sequence motif. Consequently, this causes activation of numerous downstream signaling molecules and a cascade of intracellular pathways that regulate phospholipid metabolism, arachidonate metabolism, protein phosphorylation (involving other protein kinases), calcium mobilization and transcriptional regulation. The growth-factor-dependent tyrosine kinase activity of the RTK cytoplasmic domain is the primary mechanism for generation of intracellular signals that initiate multiple cellular responses. Thus, inhibitors which would serve as alternate substrates or inhibitors of tyrosine kinase activity have the potential to block this signaling.
Many of the RTK subfamilies are recognizable on the basis of architectural similarities in the catalytic domain as well as distinctive motifs in the extracellular ligand binding regions. Based upon these structural considerations, a nomenclature defining several subfamilies of RTKs, each containing several members, has been developed (Hanks, Curr. Opin. Struc. Biol. 1:369-383, 1991; Ullrich, A., and Schlessinger, J. Cell 61:203-212, 1990). Examples of receptor subfamilies referred to on the basis of their prototypic members include: EGF-receptor, insulin receptor, platelet-derived growth factor (PDGF-receptor), fibroblast growth factor receptors (FGFRs), TRK receptor and EPH/ECK receptors. Members in each of these subfamilies represent molecular targets for the development of mimetic inhibitors that would block tyrosine kinase activity and prevent intracellular signal transduction. Several therapeutic areas in which these targets have value are identified below.
Cancer: In addition to mediating normal cellular growth, members of the EGFR family of RTKs are frequently overexpressed in a variety of aggressive epithelial carcinomas and this is thought to directly contribute to malignant tumor development. A number of studies have shown that the EGFR is frequently amplified in certain types of tumors, including glioblastomas, squamous carcinomas, and brain tumors (Wong et al., Proc. Natl. Acad Sci USA 84:6899-6903, 1987). Additionally, HER2/p185.sup.erbB2 (alternatively referred to as "neu" in the rat), HER3/p160.sup.erbB3, HER4/p180.sup.erbB4 (Plowman, G. et al., Proc. Natl. Acad. Sci. USA 90:1746-1750 (1993) are three RTKs which have extensive amino acid sequence homology to the EGFR. HER2/p185.sup.erbB2 is frequently amplified and overexpressed in human breast tumors and ovarian carcinomas (Wong et al., Proc. Natl. Acad. Sci. USA 84:6899-6903, 1987), and this amplification is correlated with poor patient prognosis. Simultaneous overexpression of p185.sup.neu and the EGFR synergistically transforms rodent fibroblasts and this condition is often observed in human cancers. Finally, HER3 expression is amplified in a variety of human adenocarcinomas. Several inhibitors are known which demonstrate inhibitory activity in vitro against the EGFR and block EGF-dependent cell proliferation which indicates therapeutic potential of compounds with this activity. In addition, in human chronic myelogenous leukemia, enhanced tyrosine kinase activity underlies the disease as a consequence of activation of the cellular c-abl protooncogene. Inhibitors would function as anticancer agents.
Angiogenesis: Currently, there are at least seven FGFR members which mediate a diverse array of biological responses, including the capacity to induce angiogenesis. In addition, a group of RTKs with seven lgLs has been proposed to represent a separate subfamily. Its known members, FLT1, FLK1 and FLT4 show a similarity of structure and expression. These receptors mediate the actions of Vascular Endothelial Growth Factor (VEGF). Several lines of evidence indicate that this subfamily of growth factor receptors play an important role in the formation of blood vessels. Since blood vessel formation is a process reactivated by tumors in order to supply oxygen to these cells, .beta.-strand mimetics that inhibit these growth factors' kinase activities could serve to suppress tumor growth through inhibition of angiogenesis.
Restenosis: The PDGF receptor is of great interest as a target for inhibition in the cardiovascular field since it is believed to play a significant role in restenosis after coronary balloon angioplasties and also in atherosclerosis. The release of PDGF by platelets at damaged surfaces of blood vessels results in stimulation of PDGF receptors on vascular smooth muscle cells, and eventual neointimal thickening. A mimetic inhibitor of kinase activity would prevent proliferation and lead to greater successful outcomes from this surgical procedure.
Many components of signal transduction pathways involve phosphorylation of serine/threonine (ser/thr) residues of protein substrates. Some of these substrates are themselves protein kinases whose activity is modulated by phosphorylation. Two prominent ser/thr-specific protein kinases play a central role in signal transduction: cyclic AMP-dependent protein kinase A (PKA) and the protein kinase C (PKC family). Numerous other serine/threonine specific kinases, including the family of mitogen-activated protein (MAP) kinases serve as important signal transduction proteins which are activated in either growth-factor receptor or cytokine receptor signaling. Other protein ser/thr kinases important for intracellular signaling are Calcium-dependent protein kinase (CaM-kinase II) and the c-raf-protooncogene.
PKC plays a crucial role in cell-surface signal transduction for controlling a variety of physiological processes (Nishizuka, Nature 334:661-665, 1988) and represents a large family of isoenzymes which differ in their structure and expression in different tissues, as well as their substrate specificity (Hug and Sarre, Biochem J. 291:329-343, 1993). Molecular cloning has demonstrated at least 8 isoenzymes. Due to this diversity and differential expression, activation of individual isoenzymes produces differing cell-specific responses: stimulation of growth, inhibition of differentiation, or induction of differentiation. Due to its ability to stimulate cellular proliferation, it represents a target for anticancer drug development (Powis, Trends in Pharm. Sci. 12:188-194, 1991). overexpression of PKC isoenzymes in mammalian cells is correlated with enhanced expression of early protooncogenes such as c-jun, c-fos, c-myc and one overexpressing cell line gives rise to tumors in nude mice.
Therapeutic applications within the area of immune regulation are evident since activation of T-cells by antigens involves activation of PKC. Activated PKC subsequently activates a branch of the signal cascade that is necessary for transcriptional activation of NF-.kappa.B, production of IL-2, and ultimately, T-cell proliferation. Inhibitors that block signaling through this branch pathway have been shown to prevent T-cell activation. Thus, mimetics that would function as inhibitors of PKC in T-cells would block signaling and serve as possible immunosuppressants useful in transplant rejection or as anticancer agents for lymphocytic leukemias. Activators of PKC cause edema and inflammation in mouse skin (Hennings et al., Carcinogenesis 8:1342-1346, 1987) and thus inhibitors are also expected to serve as potent anti-inflammatory compounds. Such anti-inflammatory activates would find use in asthma, arthritis and other inflammatory mediated processes. In addition, staurosporine and its analogs, UCN01 and CGP4125, which have been characterized as potent PKC inhibitors in vitro, have anti-tumor activity in animal models (Powis, Trends in Pharm. Sci. 12:188-194, 1991), and related compounds are being considered for clinical trials.
With regard to protease inhibition, Cathepsin B is a lysosomal cysteine protease normally involved in proenzyme processing and protein turnover. Elevated levels of activity have been implicated in tumor metastasis (Sloane, B. F. et al., "Cathepsin B and its endogenous inhibitors: the role in tumor malignancy," Cancer Metastasis Rev. 9:333-352, 1990), rheumatoid arthritis (Werb, Z. "Proteinases and matrix degradation," in Textbook of Rheumatology, Keller, W. N.; Harris, W. D.; Ruddy, S.; Sledge, C. S., Eds., 1989, W.B. Saunder Co., Philadelphia, Pa., pp. 300-321), and muscular dystrophy (Katunuma N. & Kominami E., "Abnormal expression of lysosomal cysteine proteinases in muscle wasting diseases," Rev. Physiol. Biochem. Pharmacol. 108:1-20, 1987).
Calpains are cytosolic or membrane bound Ca++-activated proteases which are responsible for degradation of cytoskeletal proteins in response to changing calcium levels within the cell. They contribute to tissue degradation in arthritis and muscular dystrophy (see Wang K.K. & Yuen P. W., "Calpain inhibition: an overview of its therapeutic potential," Trends Pharmacol. Sci. 15:412-419, 1994).
Interleukin Converting Enzyme (ICE) cleaves pro-IL-1 beta to IL-1 beta, a key mediator of inflammation, and therefore inhibitors of ICE may prove useful in the treatment of arthritis (see, e.g., Miller B. E. et al., "Inhibition of mature IL-1 beta production in murine macrophages and a murine model of inflammation by WIN 67694, an inhibitor of IL-1 beta converting enzyme," J. Immunol. 154:1331-1338, 1995). ICE or ICE-like proteases may also function in apoptosis (programmed cell death) and therefore play roles in cancer, AIDS, Alzheimer's disease, and other diseases in which disregulated apoptosis is involved (see Barr, P. J.; Tomei, L. D., "Apoptosis and its Role in Human Disease," Biotechnol. 12:487-493, 1994).
HIV protease plays a key role in the life cycle of HIV, the AIDS virus. In the final steps of viral maturation it cleaves polyprotein precursors to the functional enzymes and structural proteins of the virion core. HIV protease inhibitors were quickly identified as an excellent therapeutic target for AIDS (see Huff, J. R., "HIV protease: a novel chemotherapeutic target for AIDS," J. Med. Chem. 34:2305-2314) and have already proven useful in its treatment as evidenced by the recent FDA approval of ritonavir, Crixivan, and saquinavir.
Hepatitis C virus (HCV) is the major cause of non-A and non-B hepatitis in the world today. It is estimated to infect up to 50 million people. Currently there is no satisfactory treatment available to halt the progression of this debilitating disease. During the life cycle of the virus, a polyprotein of about 3000 amino acids is produced and is proteolytically cleaved by host and viral proteases to produce the mature viral gene products. A serine proteinase located within the HCV NS3 protein cleaves at four specific sites to produce non-structural proteins considered essential for viral replication. Hence, inhibitors of HCV protease are attractive targets for drug design, and could be of great therapeutic benefit. (Neddermann et al., Biol. Chem. 378:469-476, 1997.)
Angiotensin converting enzyme (ACE) is part of the renin-angiotensin system which plays a central role in the regulation of blood pressure. ACE cleaves angiotensin I to the octapeptide angiotensin II, a potent pressor agent due to its vasoconstrictor activity. Inhibition of ACE has proved therapeutically useful in the treatment of hypertension (Williams, G. H., "Converting-enzyme inhibitors in the treatment of hypertension," N. Engl. J. Med. 319:1517-1525, 1989.
Collagenases cleave collagen, the major constituent of the extracellular matrix (e.g., connective tissue, skin, blood vessels). Elevated collagenase activity contributes to arthritis (Krane S. M. et al., "Mechanisms of matrix degradation in rheumatoid arthritis," Ann. N.Y. Acad. Sci. 580:340-354, 1990.), tumor metastasis (Flug M. & Kopf-Maier P., "The basement membrane and its involvement in carcinoma cell invasion," Acta Anat. Basel 152:69-84, 1995), and other diseases involving the degradation of connective tissue.
Trypsin-like serine proteases form a large and highly selective family of enzymes involved in hemostasis/coagulation (Davie, E. W. and K. Fujikawa, "Basic mechanisms in blood coagulation," Ann. Rev. 799-829, 1975) and complement activation (Muller-Eberhard, H. J., "Complement," Ann. Rev. Biochem. 44:697-724, 1975). Sequencing of these proteases has shown the presence of a homologous trypsin-like core with amino acid insertions that modify specificity and which are generally responsible for interactions with other macromolecular components (Magnusson et al., "Proteolysis and Physiological Regulation," Miami Winter Symposia 11:203-239, 1976).
Thrombin, a trypsin-like serine protease, acts to provide limited proteolysis, both in the generation of fibrin from fibrinogen and the activation of the platelet receptor, and thus plays a critical role in thrombosis and hemostasis (Mann, K. G., "The assembly of blood clotting complexes on membranes," Trends Biochem. Sci. 12:229-233, 1987). Thrombin exhibits remarkable specificity in the removal of fibrinopeptides A and B of fibrinogen through the selective cleavage of only two Arg-Gly bonds of the one-hundred and eighty-one Arg- or Lys-Xaa sequences in fibrinogen (Blomback, H., Blood Clotting Enzymology, Seeger, W. H. (ed.), Academic Press, New York, 1967, pp. 143-215).
Many significant disease states are related to abnormal hemostasis, including acute coronary syndromes. Aspirin and heparin are widely used in the treatment of patients with acute coronary syndromes. However, these agents have several intrinsic limitations. For example, thrombosis complicating the rupture of atherosclerotic plaque tends to be a thrombin-mediated, platelet-dependent process that is relatively resistant to inhibition by aspirin and heparin (Fuster et al., "The pathogenesis of coronary artery disease and the acute coronary syndromes," N. Engl. J. Med. 326:242-50, 1992).
Thrombin inhibitors prevent thrombus formation at sites of vascular injury in vivo. Furthermore, since thrombin is also a potent growth factor which initiates smooth muscle cell proliferation at sites of mechanical injury in the coronary artery, inhibitors block this proliferative smooth muscle cell response and reduce restenosis. Thrombin inhibitors would also reduce the inflammatory response in vascular wall cells (Harker et al., Am. J. Cardiol. 75:12B-16B, 1995).
Furthermore, at least two well-defined transcription factors, nuclear factor (NF) .kappa.B and activator protein (AP)-1, are regulated by the intracellular reduction-oxidation (redox) state. The regulation of gene expression by the redox state holds promising therapeutic implications. For example, binding sites of the redox-regulated transcription factors NF-.kappa.B and AP-1 are located in the promoter region of a large variety of genes that are directly involved in the pathogenesis of diseases, such as AIDS, cancer, atherosclerosis and diabetic complications (Sen and Packer, FASEB Journal 10:709-720, 1996). More specifically, the binding of transcription factors such NF-.kappa.B and AP-1 to consensus sites on DNA is driven by oxidant-antioxidant homeostasis, especially by the thiol-disulfide balance.
In the case of NF-.kappa.B, a physiologically relevant thiol that plays a crucial role in the regulation of NF-.kappa.B function is reduced thioredoxin or a reduced thioredoxin-like protein. Thioredoxin is an important protein oxidoreductase with antioxidant functions. Thioredoxin has been found to upregulate DNA binding of activated NF-.kappa.B and thus augments gene expression (Schenk et al., Proc. Natl. Acad. Sci. USA 91:1672-1676, 1994). Thioredoxin has been implicated in reducing activated cytosolic NF-.kappa.B (specifically reduction of cys-62), which may thus contribute to its nuclear translocation and DNA binding (Hayashi et at., J. Biol. Chem. 268:11380-11388, 1993).
DNA binding activity of Fos and Jun in the AP-1 complex has also been found to be regulated by the redox state (Abate et al., Science 249:1157-1162, 1990). Each protein contains a single conserved cysteine (flanked by lysine and arginine) in its DNA binding domain. This thiol does not appear to be part of a disulfide bond and may exist as a sulfenic or sulfinic acid in its oxidized form. Ref-1, a bifunctional nuclear protein also possessing endonuclease DNA repair activity, stimulates AP-1 DNA binding by reduction of this regulatory cysteine. A Fos mutant in which the critical cysteine was replaced with serine elicited a three-fold increase in AP-1 DNA binding activity and was no longer subject to redox control (Okuno et al., Oncogene 8:695-701, 1993). Hence, since at least four members of the fos family, 3 of the jun family, and at least 4 of the ATF/CREB family of transcription factors all contain this conserved cysteine, redox control of transcription factors appears widespread.
As mentioned above, the regulation of transcription factors such as NF-.kappa.B and AP-1 have important therapeutic implications. For example, AP-1 is an important mediator of tumor production (Yoshioka et al., Proc. Natl. Acad. Sci. USA 92:4972-4976, 1995). Thus, compounds that repress AP-1 transcriptional activity have utility in the treatment of cancer. Furthermore, due to its direct role in regulating responses to inflammatory cytokines and endotoxins, the activation of NF-.kappa.B plays an important role in the development of chronic diseases such as rheumatoid arthritis and acute conditions such as septic shock. Autoimmune diseases, such as systemic lupus erythromatus (SLE), and Alzheimer's disease are also believed involved in activation of NF-.kappa.B. Similarly, NF-.kappa.B plays an important role in the activation of HIV gene expression. Further conditions which are believed to involve NF-.kappa.B include the flu, atherosclerosis, oncogenesis and ataxia telangiectasia (AT).
Proteins containing PDZ domains constitute an additional potential target for b-sheet mimetics. These domains of 80-100 amino acid residues mediate protein-protein interactions by binding to a consensus X-Ser/Thr-X-Val sequence at the very carboxyl terminus of proteins. There are also examples of protein interactions via PDZ domains that are internal (or non C-terminal). The crystal structure of liganded and unliganded PDZ domains have been determined and show a six b-strand and two a-helix structure that binds the consensus recognition polypeptide sequence through a b-sheet conformation. Hence, screening of appropriate b-sheet mimetics should prove a valid strategy for targeting PDZ domain-containing proteins. The targets of PDZ domain-containing proteins are varied but important in signal transduction. PSD-95, a membrane associated guanylate kinase contains three PDZ domains, two of which target the Shaker-type K.sup.+ channel and the N-methyl-D-aspartate (NMDA) receptor resulting in their clustering that is required for their function. PTPL1/FAP1, a protein tyrosine phosphatase, has five PDZ domains, two of which interact with Fas, a transmembrane protein of the tumor necrosis factor receptor family, that mediates apoptosis in many cell types. Hence, compounds targeting proteins containing the PDZ domains may prove useful as anticancer agents.
Tryptase, a trypsin-like serine protease found exclusively in mast cells, has attracted much interest due to its potential role as a mediator of inflammation. For example, in the lung tryptase is released along with other mediators of inflammation in response to binding of an inhaled antigen to cell-surface IgE receptors (Ishizaka and Ishizaka, Prog. Allergy 34:188-235, 1984). Tryptase has also been shown to cleave vasoactive intestinal peptide in vitro (Caughey et al., J. Pharmacol. Exp. Ther. 244:133-137, 1988; Tam and Caughey, Am. J. Respir. Cell Mol. Biol. 3:27-32, 1990). These results suggest that tryptase may increase bronchoconstriction via proteolysis of bronchodilating peptides in asthma patients. Consistent with this hypothesis is the recent finding that synthetic tryptase inhibitors blocked airway responses in allergic sheep (Clark et al., Am. J. Respir. Crit. Care Med. 152:2076-2083, 1995).
Tryptase activates extracellular matrix-degrading proteins prostromelysin (pro-MMP-3) and procollagenase (pro-MMP-1) via MMP-3, suggesting a role for the enzyme in tissue remodeling and inflammation (Gruber et al., J. Clin. Invest. 84:8154-8158, 1989) and therefore possibly in rheumatoid arthritis. Additionally, prostromelysin, when activated, has been shown to degrade the extracellular matrix around atherosclerotic plaques. Since abnormally high levels of tryptase-containing mast cells have been found in coronary atheromas, tryptase may play a role in atheromatous rupture (release of the thrombus), the final event of coronary atherosclerosis (Kaartinen et al., Circulation 90:1669-1678, 1994).
Other activities of tryptase include the following. Tryptase cleaves fibrinogen but is not inactivated in the presence of endogenous proteinase inhibitors (Schwartz et al., J. Immunol. 135:2762-2767, 1985; Ren et al., J. Immunol. 159:3540-3548, 1997), and may function as a local anticoagulant. It has been demonstrated to be a potent mitogen for fibroblasts and may be involved in pulmonary fibrosis and interstitial lung disease (Ruoss et al., J. Clin. Invest. 88:493-499, 1991). Tryptase may also be responsible for the activation of PAR-2 (proteinase activated receptor-2) on endothelial cells and keratinocytes (Molino et al., J. Biol. Chem. 272:4043-4049, 1997).
Given the central role of mast cells in allergic and inflammatory responses, inhibition of tryptase may result in significant therapeutic effects. Inhibitors of tryptase may be useful for preventing or treating asthma, pulmonary fibrosis and interstitial pneumonia, nephritis, hepatic fibrosis, hepatitis, hepatic cirrhosis, scleroderma, psoriasis, atopic dermatitis, chronic rheumatoid arthritis, influenza, Crohn's disease, ulcerative colitis, inflammatory bowel disease, nasal allergy, and atherosclerosis.
Chymase is a chymotrypsin-like protease that is also released from mast cells. It has been demonstrated to cleave angiotensin-I (ang-I) to angiotensin-II (ang-II) with greater efficiency and selectivity than angiotensin-I converting enzyme (ACE) (Okunishi et al., J. Hypertension 2: 227-284, 1984; Urata et al., Circ. Res. 66: 883-890, 1990). In heart tissue chymase has been shown to be a major source of ang-II production from ang-I (Dell'Italia et al., Am. J. Physiol. (Heart Circ. Physiol. 38) 269:H2065-H2073, 1996). In addition, increased chymase activity has been demonstrated in balloon-injury induced hypertrophied vessels in dogs (Shiota et al., FEBS Lett. 323:239-242, 1993). Such evidence suggests that inhibition of chymase may be therapeutic for hypertension, ischaemic heart disease, and congestive heart failure.
Urokinase-type plasminogen activator (uPA) is a trypsin-like serine proteinase which converts plasminogen to plasmin as part of the fibrinolytic system. It has long been used for thrombolysis in acute massive pulmonary embolism. Other research has shown that uPA is also a key initiator of the extra-cellular proteolytic cascade involved in cellular invasiveness (Mullins and Rohlich, Biochim. Biophys. Acta 695: 177-214, 1983; Testa and Quigly, Cancer Metast. Rev. 9:353-367, 1990). In addition uPA binds to uPA receptor (uPAR) through its growth factor domain and further modulates the activity of other proteins involved in cell migration. Overexpression of uPA appears to play a part in cancer invasiveness and metastasis; high levels of uPA, PAI-1 (plasminogen activator inhibitor-1), and uPAR correlate with poor patient prognosis. A variety of research in various model systems demonstrates that inhibitors of uPA decrease tumor cell invasiveness and metastasis (Testa and Quigly, ibid; Andreasen et al., Int J Cancer 72:1-22, 1997). Hence, inhibition of uPA may be useful in the treatment of breast cancer, prostate cancer, ovarian cancer, human renal cell cancer, gastric cancer, and lung cancer. Recent evidence indicates that inhibitors of uPA may also be useful in the prevention of restenosis (Loskutoff, Circulation 96:2772-2774, 1997).
In view of the important biological role played by the .beta.-sheet, there is a need in the art for compounds which can stabilize the intrinsic .beta.-sheet structure of a naturally occurring or synthetic peptide, protein or molecule. There is also a need in the art for making stable .beta.-sheet structures, as well as the use of such stabilized structures to effect or modify biological recognition events which involve .beta.-sheet structures. The present invention fulfills these needs and provides further related advantages.